A review into deep learning techniques for spoken language identification

Thukroo, Irshad Ahmad; Bashir, Rumaan; Giri, Kaiser J.

doi:10.1007/s11042-022-13054-0

Cited by 8 publications

(2 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recurrent Neural networks are found to be effective in learning the relevant features for the language classification [2][3] [19] . The feature extraction techniques like Log-Mel-Spectrogram, Mel-Frequency cepstral coefficients (MFCC), Perceptual Linear Prediction (PLP), Relative spectral transform Perceptual Linear Prediction (RASTA-PLP) and Linear Predictive coding (LPC) coefficients can be used for Automatic Speech Recognition (ASR) [4] .…”

Section: Related Workmentioning

confidence: 99%

Deep Learning Approaches for English-Marathi Code-Switched Detection

Bhimanwar,

Viralekar,

Anturkar

et al. 2023

ICST Transactions on Scalable Information Systems

View full text Add to dashboard Cite

During a conversation, speakers in multilingual societies frequently switch between two or more spoken languages. A linguistic action known as "code-switching" particularly alters or merges two or more languages. The development of software or tools for detecting code-switching has received very little attention. This paper proposes a Deep Learning based methods for detecting code-switched English-Marathi data. These suggested methods can be applied to various applications, including phone call merging, Intelligent AI assistants, Intelligent travelling systems to assist travellers in navigation and reservations, call centres to handle customer service issues, etc. To create a system for code switch detection, our study demonstrates a detailed analysis of extracting several audio features such as the Mel-Spectrogram, Mel-frequency Cepstral Coefficient (MFCC), and Perceptual Linear Predictive coefficients (PLP). Our team's English-Marathi code-switched dataset served as the testing ground for our methodologies. Our model's accuracy was 92.99%, with 40 MFCC coefficients having energy coefficient serving as the zeroth coefficient.

show abstract

Section: Related Workmentioning

confidence: 99%

Deep Learning Approaches for English-Marathi Code-Switched Detection

Bhimanwar,

Viralekar,

Anturkar

et al. 2023

ICST Transactions on Scalable Information Systems

View full text Add to dashboard Cite

show abstract

“…Recent decades, the efficiency and effectiveness of the machine leaning (ML) and deep learning (DL) approaches have been proven in many fields such as language identification (LID) [5], emotion speech recognition [6], detection of COVID-19 [7]. Therefore, recently, extensive researches have been conducted by utilizing ML and DL algorithms in the detection of PD based on hand-draw spiral images [8]- [10].…”

Section: Introductionmentioning

confidence: 99%

Convolutional neural network for the detection of Parkinson disease based on hand-draw spiral images

Alniemi

Mahmood

2023

IJEECS

View full text Add to dashboard Cite

Parkinson’s disease (PD) is a chronic and increasing sickness that hits hundreds-thousands of people globally. Patients who are infected by PD have been proven to show some common symptoms such as slowness of movement, tremors, and freezing of gait. One of the most popular exams to detect the PD is to use the handwritten assessment tool, where the individuals are asked to draw spirals on a template paper. Therefore, this study proposes a convolutional neural network algorithm for detecting the PD by utilizing the hand-draw spiral images. In the present study, balanced spiral images dataset has been utilized for both categories (i.e., Parkinson and healthy). The dataset contains 102 samples as a total number of spiral images (i.e., 51 Parkinson and 51 healthy). Moreover, numerous evaluation measurements were utilized in order to assess the proposed approach such as recall, precision, accuracy, F-measure, specificity, Matthew's correlation coefficient (MCC), and G-mean. Based on the outcomes of the experiments, the proposed approach achieves 93.33% accuracy, 86.67% specificity, 88.24% precision, 100.00% recall, 93.75% F-measure, 93.93% G-mean, and 87.45% MCC. The proposed approach demonstrates promising outcomes in the detection of PD. As well as the proposed convolutional neural network (CNN) approach was outperformed all its comparatives regarding the classification accuracy rate.

show abstract